Skip to content

Make DefId repr(C), optimize big-endian field order #92012

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Jan 11, 2022
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 8 additions & 1 deletion compiler/rustc_span/src/def_id.rs
Original file line number Diff line number Diff line change
Expand Up @@ -221,10 +221,17 @@ impl<D: Decoder> Decodable<D> for DefIndex {
#[derive(Clone, PartialEq, Eq, PartialOrd, Ord, Copy)]
// On below-64 bit systems we can simply use the derived `Hash` impl
#[cfg_attr(not(target_pointer_width = "64"), derive(Hash))]
// Note that the order is essential here, see below why
#[repr(C)]
// We guarantee field order. Note that the order is essential here, see below why.
pub struct DefId {
// cfg-ing the order of fields so that the `DefIndex` which is high entropy always ends up in
// the lower bits no matter the endianness. This allows the compiler to turn that `Hash` impl
// into a direct call to 'u64::hash(_)`.
#[cfg(not(all(target_pointer_width = "64", target_endian = "big")))]
pub index: DefIndex,
pub krate: CrateNum,
#[cfg(all(target_pointer_width = "64", target_endian = "big"))]
pub index: DefIndex,
}

// On 64-bit systems, we can hash the whole `DefId` as one `u64` instead of two `u32`s. This
Expand Down